Disposal of high-level and/or long-lived radioactive waste in engineered facilities, or repositories, located underground in suitable geological formations, is being widely investigated world-wide as a long-term management solution. This is in order to protect humans and the environment, both now and in the future. From a quantitative point of view, a repository is said to be safe if it meets the relevant safety standards, such as internationally recommended or specified by the responsible national regulatory authorities. In recent years the scope of the safety assessment has broadened to include the collation of a broader range of evidence and arguments that complement and support the reliability of the results of quantitative analyses. The broader term “post-closure safety case”, or simply “safety case”,1 is used to refer to these studies. It has also become evident that repository development will involve a number of step-by-step stages, punctuated by interdependent decision making on whether and how to move from one stage to the subsequent one. These decisions require a clear and traceable presentation of robust technical arguments that will help to give confidence in the feasibility and safety of a proposed concept. The depth of understanding and technical information available to support decisions will increase from step to step. The safety case is a key input to support a decision to move to the next stage in repository development. It reflects the state of understanding and the results of the research and development (R&D) undertaken at a certain stage, and supports decisions concerning future R&D efforts. 

This workshop is the second in a series dealing with geosphere stability for various host rock types (i.e. crystalline rocks, argillaceous rocks and evaporites). It focused on crystalline rock – a term that, in this context, is meant to include all types of hard, fractured rock – i.e. those not otherwise planned to be covered by the other workshops in the series focused on clay or salt environments. In particular, this workshop was designed to include the various hard rocks being investigated by the USA and Japan as potential host environments for geological disposal, e.g. tuffs. An important objective of the overall “geosphere stability” initiative under the NEA IGSC is to ensure that the views of the broader scientific community are taken into account in developing an understanding of geosphere stability that can be applied to the management of radioactive waste. 

The workshop was divided into four plenary sessions and a poster session. The first plenary session discussed the general framework to be followed when considering deep disposal in crystalline rocks. The subsequent sessions focused on more specific processes of importance to the stability of crystalline rocks, their response and resilience to natural perturbations and arguments to support confidence in their stability. 

J. Andersson et al. (Streamflow AB, Posiva, REC, VTT and Pöyry Environment) presented the discipline-integrated approach to modelling that is being used for describing the Olkiluoto site in Finland. The Olkiluoto Modelling Task Force (OMTF) has been established for planning and integrating the results and the modelling work in the different supporting disciplines. The main duty of the OMTF is to develop site descriptive models of the Olkiluoto site, as well as predicting and evaluating the disturbance created by construction of the ONKALO ramp and the characterisation tunnels. The resulting geosynthesis is reported in a series of Site Reports. Future updates of the Site Report will form part of the Safety Case portfolio, supporting a licence application for constructing a repository for Finland’s spent nuclear fuel at the site. 

The workshop ended with an extensive period of discussion, followed by a presentation and discussion of the key concluding messages. The discussion period was initially divided into the subjects covered by the four sessions. 

A number of detailed safety functions for a granitic host rock, subordinate to the main functions containment and retardation, are presented, based on SKB’s safety assessment SR-Can for the KBS-3 concept. The host rock should provide a favourable environment for the repository from the point of view of chemical, hydraulic and radionuclide transport, mechanical and thermal properties, which are further specified as a number of subordinate safety functions. These are strongly linked to the functions and properties of the canister and the clay buffer in the KBS-3 concept. Issues related to geosphere stability during the one million year assessment period are discussed and primarily relate to perturbations caused by future colder and dryer climates yielding glacial and/or permafrost conditions and the impact these perturbations may have on the safety functions. In the SR-Can assessment, it was concluded that the granitic host rocks at the analysed sites provide a sufficiently favourable and stable environment for the vast majority of the 6 000 analysed deposition holes. The residual radiological risks associated with the repository, as pessimistically calculated based on preliminary understanding of the two candidate sites, are, from the point of view of the host rock, sensitive to details in the repository layout. In particular, it is important to be able to avoid deposition holes intersected by large or highly transmissive and hydraulically connected fractures and thus to understand and being able to quantify the heterogeneous character of the fractured host rock, in particular at repository depth. A number of issues related to geosphere stability where improved knowledge could lead to more realistic assessments are also identified.

We define the structure of a methodology being developed by NUMO to assess the likelihood that volcanic or rock deformation processes could significantly affect sites that may emerge from the Japanese volunteer siting process. Simple criteria are used to exclude clearly unsuitable areas at the outset: the present methodology is developing probabilistic techniques to look at likelihood in nonexcluded areas. These are based on scientific models for the occurrence of volcanism and a range of proxy indicators for potential future rock deformation.

The Olkiluoto site has been chosen as a repository site for the high-level nuclear waste in 2001. Investigations in the site have been ongoing since 1987. The basic idea in the crystalline nuclear waste site still is that the solid repository block surrounded by deformation zones can host a safe repository. It is impossible to say that neither the major ductile nor large-scale brittle deformation zones are stable, but it is possible to say that the tectonic processes have been active in a stable way for billions of years by reactivating the old features time after time and there are no signs of new large features formed in the vicinity of the site during the present time including postglacial period. Understanding the geological history, especially the ductile deformation and over thrusting, begins from the understanding of the lithological features, mainly rock types, in the island. Vice versa, the occurrence and location of the lithological features are interpreted according to ductile deformation. In addition, you cannot study only present brittle deformation but you need to understand also older ductile and lithological features to be able to understand why these brittle features are where they are and to be able to predict them. 

The present paper describes a method to calculate fracture shear displacements occurring as a result of the stress waves and the static stress redistribution following a seismic slip on a nearby fault. Such secondary fracture shear displacements can theoretically, if large enough, damage intersected canisters containing spent nuclear fuel. The method is applied to a type of seismic event that is, or may be, of concern for the long term safety of a KBS-3 nuclear fuel repository: namely, endglacial earthquakes of magnitude 6 and larger. The numerical scheme used to simulate rupture initiation and propagation is described and illustrated. Result examples are given that show that secondary shear displacements on 300 m diameter fractures will be smaller than the damage criterion now applied by the Swedish Nuclear Fuel and Waste Management Co (SKB), provided that the fracture center is at a distance of 200 m or larger from the fault plane. The relevance of the rupture representation is discussed based on comparison with slip velocity records from a recent, large and well-documented crustal earthquake and with stress drop estimates made for large endglacial faults observed in Lapland, northern Sweden. 

Posiva, the nuclear waste management organization in Finland, is currently constructing an underground research facility at the Olkiluoto site. The Olkiluoto Modelling Task Force (OMTF) has been established for planning and integrating the results and the modelling work in the different supporting disciplines. The main duty of the OMTF is to develop site descriptive models of the Olkiluoto site, as well as predicting and evaluating the disturbance created by construction of the ONKALO ramp and characterization tunnels. Various approaches are utilised including crossdiscipline workshops to ensure a common basis for modelling; prediction/outcome studies; and carrying out an overall confidence assessment are key synthesizing and integrating tools. The resulting geosynthesis is reported in a series of Site Reports. Future updates of the Site Report will be part of the Safety Case portfolio supporting a licence application for constructing a final repository for Finland’s spent nuclear fuel at the site.